As technology advances, more and more portable electronic devices, such as mobile phones, digital cameras, tablet computers, laptops, and the like, have become necessary tools for people in life. Since the portable electronic device is commonly for personal use and has a certain of privacy, data stored therein, such as a telephone book, photos, personal information, and the like, are owned privately. Once the electronic device is lost, these data may be utilized by others, resulting in unnecessary losses. Although there is a way of utilizing password protection to avoid the electronic device from being used by others currently, the password is easily leaked and cracked, and has lower security. In addition, a user needs to remember the password to use the electronic device, and if the password is forgotten, it will bring the user a lot of inconvenience. Therefore, a way of utilizing a personal fingerprint identification system is developed currently to achieve an objective of identity authentication so as to improve data security.
Generally, a touch from a finger is received with a pixel array circuit in a fingerprint identification system, and the pixel array circuit converts a capacitance between the pixel array circuit and the finger into a pixel output signal. Affected by a noise, the pixel output signal does not possess good statistical linearity, which is not conducive to back-end image processing. Therefore, how to improve statistical linearity of an output signal in a fingerprint identification system has become one of goals of the industry.